Analysing and Visualizing the Indian Movies dataset

The following code visualize the IMDB dataset of indian movies. The code is divided into four sections:

1. Importing Dataset
2. Analysing Dataset
3. Cleaning Data
4. Plotting Data

Importing all required libraries

# import libraries

import numpy as np
import pandas as pd

import seaborn as sns
from matplotlib import pyplot as plt
import plotly.graph_objects as go
import plotly.express as px

import collections
from collections import namedtuple

import pandas_profiling
from IPython.display import display

import warnings
warnings.filterwarnings('ignore')

Importing Data

# Read Data

df = pd.read_csv("indian_movies.csv",encoding="ISO-8859-1")
df.head()

	Name	Year	Duration	Genre	Rating	Votes	Director	Actor 1	Actor 2	Actor 3
0		NaN	NaN	Drama	NaN	NaN	J.S. Randhawa	Manmauji	Birbal	Rajendra Bhatia
1	#Gadhvi (He thought he was Gandhi)	(2019)	109 min	Drama	7.0	8	Gaurav Bakshi	Rasika Dugal	Vivek Ghamande	Arvind Jangid
2	#Homecoming	(2021)	90 min	Drama, Musical	NaN	NaN	Soumyajit Majumdar	Sayani Gupta	Plabita Borthakur	Roy Angana
3	#Yaaram	(2019)	110 min	Comedy, Romance	4.4	35	Ovais Khan	Prateik	Ishita Raj	Siddhant Kapoor
4	...And Once Again	(2010)	105 min	Drama	NaN	NaN	Amol Palekar	Rajat Kapoor	Rituparna Sengupta	Antara Mali

Analysing Dataset

# Analysing Data
report = pandas_profiling.ProfileReport(df)
report.to_file(output_file='profilingReport.html')

Cleaning Data

After looking at the Pandas Profiling Report we come to a following conclusion:

1. There are multiple duplicate values in the dataset which needs to be removed.
2. There are many missing values in the dataset and if a movie does not have the required rating or year or director name then it will hard to the visualise the data. Hence, such rows needs to be eliminated.
3. The year field has brackets which are not required and the movie duration field has 'min' at the end. This has to be removed so we have datatype 'int' and not ''string'.
4. There are multiple entries in Genre column seperate by a ' , ' . Spliting of data into multiple columns is required for such data.

# Missing Values
null_rows = df[df.iloc[: , 1:9].isna().apply(lambda x: all(x), axis=1)]
print("Missing values from each column:")
null_rows.head()

Missing values from each column:

	Name	Year	Duration	Genre	Rating	Votes	Director	Actor 1	Actor 2	Actor 3
1836	Bang Bang Reloaded	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
1920	Battle of bittora	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2653	Campus	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
3403	Dancing Dad	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
3807	Dial 100	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

df = df[~df.iloc[: , 1:9].isna().apply(lambda x: all(x), axis=1)]
df.shape

(15501, 10)

# Duplicate Values
duplicate = df[df.duplicated(subset = ['Name', 'Year'], keep = False)]
print("Duplicate rows according to Name and Year:")
duplicate.head()

Duplicate rows according to Name and Year:

	Name	Year	Duration	Genre	Rating	Votes	Director	Actor 1	Actor 2	Actor 3
135	A Ballad of Maladies	(2016)	86 min	Documentary, Music	NaN	NaN	Sarvnik Kaur	Tushar Madhav	NaN	NaN
136	A Ballad of Maladies	(2016)	86 min	Documentary, History, Music	NaN	NaN	Sarvnik Kaur	Tushar Madhav	NaN	NaN
585	Achanak	(1988)	NaN	Thriller	NaN	NaN	Rakesh Kashyap	Om Puri	Kunickaa Sadanand	Kanwaljit Singh
586	Achanak	(1988)	NaN	Drama	NaN	NaN	Kavi Raz	Alka	Amarjeet Kaur	Kavi Raz
953	Amrit	(1941)	153 min	NaN	NaN	NaN	Master Vinayak	Dada Salvi	Baburao Pendharkar	Lalita Pawar

# Droping Duplicates
df.drop_duplicates(subset=['Name', 'Year'], inplace=True)
df.shape

(15485, 10)

null_rows = df[df.iloc[: , [1,2,4,5]].isna().apply(lambda x: all(x), axis=1)]
df = df[~df.iloc[: , [1,2,4,5]].isna().apply(lambda x: all(x), axis=1)]
df['Year'] = df['Year'].str.replace(r'(', '').str.replace(r')', '')   # Removing brackets from Year
df['Duration'] = df['Duration'].str.replace(r' min', '')              # Removing min from Duration
df.drop(df.loc[df['Year']=='2022'].index, inplace = True)             # Removing 2022 movies data     
df.head()

	Name	Year	Duration	Genre	Rating	Votes	Director	Actor 1	Actor 2	Actor 3
1	#Gadhvi (He thought he was Gandhi)	2019	109	Drama	7.0	8	Gaurav Bakshi	Rasika Dugal	Vivek Ghamande	Arvind Jangid
2	#Homecoming	2021	90	Drama, Musical	NaN	NaN	Soumyajit Majumdar	Sayani Gupta	Plabita Borthakur	Roy Angana
3	#Yaaram	2019	110	Comedy, Romance	4.4	35	Ovais Khan	Prateik	Ishita Raj	Siddhant Kapoor
4	...And Once Again	2010	105	Drama	NaN	NaN	Amol Palekar	Rajat Kapoor	Rituparna Sengupta	Antara Mali
5	...Aur Pyaar Ho Gaya	1997	147	Comedy, Drama, Musical	4.7	827	Rahul Rawail	Bobby Deol	Aishwarya Rai Bachchan	Shammi Kapoor

Plotting Data

Plotting various graphs to visualize the data throughout the years.

Number of movies by year of launch

year_count=df.groupby('Year').apply(lambda x:x['Name'].count()).reset_index(name='Count')
fig = px.bar(year_count, y='Count', x='Year', text='Count',title='Number of movies by year of launch')
fig.update_traces(texttemplate='%{text:.2s}', textposition='outside')
fig.update_layout(uniformtext_mode='show', xaxis = dict (title = 'Year of Movie Release', titlefont_size=16), yaxis=dict(title='Count of Movies Released', titlefont_size=16,tickfont_size=14))
fig.show()

Genres through the Years

dummies = df['Genre'].str.get_dummies(sep=', ')
df_genre = pd.concat([df, dummies], axis=1)
df_genre = df_genre.replace(0, np.nan)
df_genre.drop(['Name','Year','Duration','Genre','Rating','Votes','Director','Actor 1','Actor 2','Actor 3'], axis=1).sum().sort_values(ascending=False)

Drama          7064.0
Action         3479.0
Romance        2399.0
Comedy         1955.0
Thriller       1643.0
Crime          1290.0
Family          930.0
Musical         580.0
Adventure       524.0
Mystery         501.0
Horror          499.0
Fantasy         448.0
Documentary     377.0
History         195.0
Biography       192.0
Animation       120.0
Music            85.0
Sport            66.0
Sci-Fi           55.0
War              43.0
News              9.0
Western           5.0
Reality-TV        3.0
Short             1.0
dtype: float64

df_genre_count = df_genre.drop(['Name','Duration','Genre','Rating','Votes','Director','Actor 1','Actor 2','Actor 3'], axis=1)
df_genre_count = df_genre_count.groupby('Year').sum()
df_genre_count.reset_index(level=0, inplace=True)

fig = go.Figure()
fig.add_trace(go.Scatter(x=df_genre_count.Year, y=df_genre_count.Action,
                    mode='lines',
                    name='Action',
                    marker_color = '#2E91E5'))
fig.add_trace(go.Scatter(x=df_genre_count.Year, y=df_genre_count.Adventure,
                    mode='lines',
                    name='Adventure',
                    marker_color = '#E15F99'))
fig.add_trace(go.Scatter(x=df_genre_count.Year, y=df_genre_count.Comedy,
                    mode='lines',
                    name='Comedy',
                    marker_color = '#DA16FF'))
fig.add_trace(go.Scatter(x=df_genre_count.Year, y=df_genre_count.Drama,
                    mode='lines',
                    name='Drama',
                    marker_color = '#750D86'))
fig.add_trace(go.Scatter(x=df_genre_count.Year, y=df_genre_count.Horror,
                    mode='lines',
                    name='Horror',
                    marker_color = '#FB00D1'))
fig.add_trace(go.Scatter(x=df_genre_count.Year, y=df_genre_count.Romance,
                    mode='lines',
                    name='Romance',
                    marker_color = '#A777F1'))
fig.add_trace(go.Scatter(x=df_genre_count.Year, y=df_genre_count.Thriller,
                    mode='lines',
                    name='Thriller',
                    marker_color = '#6C4516'))


fig.update_layout(
    title='Genre through the Years',
    xaxis_tickfont_size=14,
        height = 800,
    yaxis=dict(
        title='Count',
        titlefont_size=16,
        tickfont_size=14,
    ),
    legend=dict(
        y=0,
        x=1.0,
        bgcolor='rgba(255, 255, 255, 0)',
        bordercolor='rgba(255, 255, 255, 0)'
))
fig.show()

Top 20 actors by the number of movies made

Actor1 = df[['Actor 1','Year']].rename(columns = {'Actor 1': 'Actor'}, inplace = False)
Actor2 = df[['Actor 2','Year']].rename(columns = {'Actor 2': 'Actor'}, inplace = False)
Actor3 = df[['Actor 3','Year']].rename(columns = {'Actor 3': 'Actor'}, inplace = False)
Actor_Year = pd.concat([Actor1, Actor2, Actor3], ignore_index=True).dropna()
Actor_Year['Count'] = 1
Actor_Top = Actor_Year['Actor'].value_counts().rename_axis('Actor').reset_index(name='Count')

fig = px.bar(Actor_Top[0:20],
             x='Actor',
             y='Count',
             title = 'Top 20 actors by number of Movies made')
fig.show()

Top 20 actors by number of movies made through the year

Top_Actor = Actor_Top[0:20]['Actor']
Actor_Year = Actor_Year[Actor_Year['Actor'].isin(Top_Actor)].set_index('Actor').loc[Top_Actor].reset_index(level=0)
Actor_Year = Actor_Year.sort_values('Year')
Actor_Year['Year'] =  pd.to_datetime(Actor_Year['Year'])

fig = px.strip(Actor_Year,
               x="Year",
               y='Actor',
               color = "Actor",
               color_discrete_sequence=px.colors.qualitative.Dark24,
               height = 600,
               title = 'Top 20 actors by number of Movies made through the Years')
fig.show()